Have you noticed a player's rating jumps after a big upset, but it barely moves when they beat someone they were always going to beat? That's that's the Glicko rating system in action. Once you understand how it works, those numbers stop feeling random and start telling a real story about who belongs where in your draw.
Why Elo falls short for club play
The Elo system — the original chess rating system — gives every player a single number and shifts it after each match. Simple, but it has a critical blind spot for club environments: it treats every rating as equally trustworthy.
A player who competes every week has a well-tested rating. A player returning after three months off does not. Elo can't tell the difference between these players, but Glicko can.
Two numbers, not one
The Glicko system, developed by statistician Mark Glickman, tracks two values for every player: their rating (a measure of skill) and their rating deviation (RD) (a measure of confidence in that rating).
A low RD means the system is confident. A high RD means the player hasn't competed enough recently for the rating to be pinned down.
Think of RD as a margin of error. A player rated 1500 with an RD of 50 is almost certainly between 1450 and 1550. The same rating with an RD of 200 could be anywhere from 1300 to 1700. Same number — very different story.
How Glicko updates after every match
When a result is recorded, Glicko weighs three things: the gap between the two players' ratings, the RD of the opponent, and who won.
Beat a player whose rating is rock solid and the winner's score climbs meaningfully. Beat a player whose RD is sky-high — who might be far better or worse than their number suggests — and the system adjusts conservatively. One upset against someone of unknown skill level shouldn't skew anyone's rating.
This is what makes Glicko fairer than simpler systems like Elo: it rewards consistency.
What it means for your club's players
In practice, here's what Glicko looks like across a season:
- New players start with a high RD — their rating moves fast in early weeks and stabilises as results accumulate
- Returning players see their RD creep up during absence, so their first few matches back carry more weight
- Regular competitors develop tight, reliable ratings that hold up against scrutiny
The ratings earn their accuracy. They're not handed out — they're built through play.
Better ratings mean better draws
Accurate ratings aren't just interesting numbers. They're the foundation for fair groupings. When players are matched against opponents at a similar level, every match is competitive — close enough that results could go either way. That's what makes a session worth showing up for.
Table Tennis Now calculates Glicko ratings automatically after every event. No spreadsheet. No manual calculation. No waiting until next week.
The longer your club plays, the sharper the picture
A club with three sessions of data has a rough map. A club with six months of results has a detailed one. Players can see exactly where they stand, track how far they've come, and know that the number next to their name reflects real, tested performance.
That's the kind of competitive environment that keeps players coming back — and keeps growing a club.
Ready to put Glicko to work for your club? Table Tennis Now handles the ratings, draws, and leaderboards — so you can focus on the matches.